Paper — every trading figure here (positions, orders, P&L, equity, returns) is simulated on an Alpaca paper account. No real capital is at risk, and no order is ever placed with your money. Running costs shown under ops, such as LLM spend, are real.
Strategy Optimization
Every trade decision feeds an evaluation loop: the system scores its own reasoning and learns from your feedback. Applying what it learns stays gated — parameters only change once the evidence says they should, and you approve.
Where the loop stands today
Reading the loop’s current state…
Annotations (30d)
Judge runs (7d)
Pending proposals
Applied revisions
Judge score (7d mean)
The Loop — how each cycle makes the next one sharper
Loop armed — the pulse rotates clockwise to signal each stage is wired and listening. Real counts replace this idle animation as soon as the first decision flows through.
Revision Timeline
Every parameter change the system has proposed — including those you declined. The audit trail shows not just what changed, but what was considered and why.
Evaluation & Proposals
Judge verdicts score decision quality. Proposals surface when patterns emerge. You approve; the system applies.
LLM-Judge Verdicts
Proposal Queue
Impact Preview
Click "Preview" on any proposal to see before/after performance estimates — win rate, Sharpe ratio, max drawdown, and trade count compared to the current strategy.